Convolutional Neural Networks

Project: Write an Algorithm for a Dog Identification App


In this notebook, some template code has already been provided for you, and you will need to implement additional functionality to successfully complete this project. You will not need to modify the included code beyond what is requested. Sections that begin with '(IMPLEMENTATION)' in the header indicate that the following block of code will require additional functionality which you must provide. Instructions will be provided for each section, and the specifics of the implementation are marked in the code block with a 'TODO' statement. Please be sure to read the instructions carefully!

Note: Once you have completed all of the code implementations, you need to finalize your work by exporting the Jupyter Notebook as an HTML document. Before exporting the notebook to html, all of the code cells need to have been run so that reviewers can see the final implementation and output. You can then export the notebook by using the menu above and navigating to File -> Download as -> HTML (.html). Include the finished document along with this notebook as your submission.

In addition to implementing code, there will be questions that you must answer which relate to the project and your implementation. Each section where you will answer a question is preceded by a 'Question X' header. Carefully read each question and provide thorough answers in the following text boxes that begin with 'Answer:'. Your project submission will be evaluated based on your answers to each of the questions and the implementation you provide.

Note: Code and Markdown cells can be executed using the Shift + Enter keyboard shortcut. Markdown cells can be edited by double-clicking the cell to enter edit mode.

The rubric contains optional "Stand Out Suggestions" for enhancing the project beyond the minimum requirements. If you decide to pursue the "Stand Out Suggestions", you should include the code in this Jupyter notebook.


Why We're Here

In this notebook, you will make the first steps towards developing an algorithm that could be used as part of a mobile or web app. At the end of this project, your code will accept any user-supplied image as input. If a dog is detected in the image, it will provide an estimate of the dog's breed. If a human is detected, it will provide an estimate of the dog breed that is most resembling. The image below displays potential sample output of your finished project (... but we expect that each student's algorithm will behave differently!).

Sample Dog Output

In this real-world setting, you will need to piece together a series of models to perform different tasks; for instance, the algorithm that detects humans in an image will be different from the CNN that infers dog breed. There are many points of possible failure, and no perfect algorithm exists. Your imperfect solution will nonetheless create a fun user experience!

The Road Ahead

We break the notebook into separate steps. Feel free to use the links below to navigate the notebook.

  • Step 0: Import Datasets
  • Step 1: Detect Humans
  • Step 2: Detect Dogs
  • Step 3: Create a CNN to Classify Dog Breeds (from Scratch)
  • Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)
  • Step 5: Write your Algorithm
  • Step 6: Test Your Algorithm

Step 0: Import Datasets

Make sure that you've downloaded the required human and dog datasets:

  • Download the dog dataset. Unzip the folder and place it in this project's home directory, at the location /dog_images.

  • Download the human dataset. Unzip the folder and place it in the home directory, at location /lfw.

Note: If you are using a Windows machine, you are encouraged to use 7zip to extract the folder.

In the code cell below, we save the file paths for both the human (LFW) dataset and dog dataset in the numpy arrays human_files and dog_files.

In [1]:
import numpy as np
from glob import glob

# load filenames for human and dog images
human_files = np.array(glob("/data/lfw/*/*"))
dog_files = np.array(glob("/data/dog_images/*/*/*"))

# print number of images in each dataset
print('There are %d total human images.' % len(human_files))
print('There are %d total dog images.' % len(dog_files))
There are 13233 total human images.
There are 8351 total dog images.

Step 1: Detect Humans

In this section, we use OpenCV's implementation of Haar feature-based cascade classifiers to detect human faces in images.

OpenCV provides many pre-trained face detectors, stored as XML files on github. We have downloaded one of these detectors and stored it in the haarcascades directory. In the next code cell, we demonstrate how to use this detector to find human faces in a sample image.

In [2]:
import cv2                
import matplotlib.pyplot as plt                        
%matplotlib inline                               

# extract pre-trained face detector
face_cascade = cv2.CascadeClassifier('haarcascades/haarcascade_frontalface_alt.xml')

# load color (BGR) image
img = cv2.imread(human_files[0])
# convert BGR image to grayscale
gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)

# find faces in image
faces = face_cascade.detectMultiScale(gray)

# print number of faces detected in the image
print('Number of faces detected:', len(faces))

# get bounding box for each detected face
for (x,y,w,h) in faces:
    # add bounding box to color image
    cv2.rectangle(img,(x,y),(x+w,y+h),(255,0,0),2)
    
# convert BGR image to RGB for plotting
cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)

# display the image, along with bounding box
plt.imshow(cv_rgb)
plt.show()
Number of faces detected: 1

Before using any of the face detectors, it is standard procedure to convert the images to grayscale. The detectMultiScale function executes the classifier stored in face_cascade and takes the grayscale image as a parameter.

In the above code, faces is a numpy array of detected faces, where each row corresponds to a detected face. Each detected face is a 1D array with four entries that specifies the bounding box of the detected face. The first two entries in the array (extracted in the above code as x and y) specify the horizontal and vertical positions of the top left corner of the bounding box. The last two entries in the array (extracted here as w and h) specify the width and height of the box.

Write a Human Face Detector

We can use this procedure to write a function that returns True if a human face is detected in an image and False otherwise. This function, aptly named face_detector, takes a string-valued file path to an image as input and appears in the code block below.

In [3]:
# returns "True" if face is detected in image stored at img_path
def face_detector(img_path):
    img = cv2.imread(img_path)
    gray = cv2.cvtColor(img, cv2.COLOR_BGR2GRAY)
    faces = face_cascade.detectMultiScale(gray)
    return len(faces) > 0

(IMPLEMENTATION) Assess the Human Face Detector

Question 1: Use the code cell below to test the performance of the face_detector function.

  • What percentage of the first 100 images in human_files have a detected human face?
  • What percentage of the first 100 images in dog_files have a detected human face?

Ideally, we would like 100% of human images with a detected face and 0% of dog images with a detected face. You will see that our algorithm falls short of this goal, but still gives acceptable performance. We extract the file paths for the first 100 images from each of the datasets and store them in the numpy arrays human_files_short and dog_files_short.

Answer: Human Files detected 98% faces Dog Files detected 17% faces

In [6]:
from tqdm import tqdm

human_files_short = human_files[:100]
dog_files_short = dog_files[:100]

#-#-# Do NOT modify the code above this line. #-#-#

## TODO: Test the performance of the face_detector algorithm 
## on the images in human_files_short and dog_files_short.
h=0
d=0
percent = ([face_detector(x) for x in human_files_short].count(True))/len(human_files_short)*100
print(f"Human face detected in Human Images : {percent} %")

percent = ([face_detector(x) for x in dog_files_short].count(True))/len(dog_files_short)*100
print(f"Dog detected as Human Face : {percent} %") 
Human face detected in Human Images : 98.0 %
Dog detected as Human Face : 17.0 %

We suggest the face detector from OpenCV as a potential way to detect human images in your algorithm, but you are free to explore other approaches, especially approaches that make use of deep learning :). Please use the code cell below to design and test your own face detection algorithm. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [5]:
### (Optional) 
### TODO: Test performance of anotherface detection algorithm.
### Feel free to use as many code cells as needed.

Step 2: Detect Dogs

In this section, we use a pre-trained model to detect dogs in images.

Obtain Pre-trained VGG-16 Model

The code cell below downloads the VGG-16 model, along with weights that have been trained on ImageNet, a very large, very popular dataset used for image classification and other vision tasks. ImageNet contains over 10 million URLs, each linking to an image containing an object from one of 1000 categories.

In [4]:
import torch
import torchvision.models as models

# define VGG16 model
VGG16 = models.vgg16(pretrained=True)

# check if CUDA is available
use_cuda = torch.cuda.is_available()

# move model to GPU if CUDA is available
if use_cuda:
    VGG16 = VGG16.cuda()
Downloading: "https://download.pytorch.org/models/vgg16-397923af.pth" to /root/.torch/models/vgg16-397923af.pth
100%|██████████| 553433881/553433881 [00:11<00:00, 50048715.62it/s]

Given an image, this pre-trained VGG-16 model returns a prediction (derived from the 1000 possible categories in ImageNet) for the object that is contained in the image.

(IMPLEMENTATION) Making Predictions with a Pre-trained Model

In the next code cell, you will write a function that accepts a path to an image (such as 'dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg') as input and returns the index corresponding to the ImageNet class that is predicted by the pre-trained VGG-16 model. The output should always be an integer between 0 and 999, inclusive.

Before writing the function, make sure that you take the time to learn how to appropriately pre-process tensors for pre-trained models in the PyTorch documentation.

In [5]:
from PIL import Image
import os
import numpy as np
import torch

import torchvision
from torchvision import datasets, models, transforms
import matplotlib.pyplot as plt

%matplotlib inline

def VGG16_predict(img_path):
    '''
    Use pre-trained VGG-16 model to obtain index corresponding to 
    predicted ImageNet class for image at specified path
    
    Args:
        img_path: path to an image
        
    Returns:
        Index corresponding to VGG-16 model's prediction
    '''
    
    ## TODO: Complete the function.
    ## Load and pre-process an image from the given img_path
    ## Return the *index* of the predicted class for that image
    #img_path ='dogImages/train/001.Affenpinscher/Affenpinscher_00001.jpg'
    data_transform = transforms.Compose([transforms.Resize(size=(244, 244)),
                                         transforms.ToTensor(),
                                         transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                                             std=[0.229, 0.224, 0.225])])
    
    img_pil = Image.open(img_path).convert('RGB')
    img = data_transform(img_pil).float()
    #img = Variable(img)
    img = img.unsqueeze(0)
    VGG16.eval()
    if use_cuda:
        img = img.cuda()
    ret = VGG16(img)
    return torch.max(ret,1)[1].item()
    
In [28]:
VGG16_predict(dog_files_short[0])
Out[28]:
243

(IMPLEMENTATION) Write a Dog Detector

While looking at the dictionary, you will notice that the categories corresponding to dogs appear in an uninterrupted sequence and correspond to dictionary keys 151-268, inclusive, to include all categories from 'Chihuahua' to 'Mexican hairless'. Thus, in order to check to see if an image is predicted to contain a dog by the pre-trained VGG-16 model, we need only check if the pre-trained model predicts an index between 151 and 268 (inclusive).

Use these ideas to complete the dog_detector function below, which returns True if a dog is detected in an image (and False if not).

In [6]:
### returns "True" if a dog is detected in the image stored at img_path
def dog_detector(img_path):
    ## TODO: Complete the function.
    output = VGG16_predict(img_path)
    if output>=151 and output<=268:
        return True 
    return False # true/false
In [10]:
dog_detector(dog_files_short[1])
Out[10]:
True
In [11]:
dog_detector(human_files_short[3])
Out[11]:
False

(IMPLEMENTATION) Assess the Dog Detector

Question 2: Use the code cell below to test the performance of your dog_detector function.

  • What percentage of the images in human_files_short have a detected dog?
  • What percentage of the images in dog_files_short have a detected dog?

Answer: 1) 0% of dogs were dectected in human files 2) 100% dog images were detected in dog files

In [12]:
### TODO: Test the performance of the dog_detector function
### on the images in human_files_short and dog_files_short.
h=0
d=0
percent = ([dog_detector(x) for x in human_files_short].count(True))/len(human_files_short)*100
print(f"Human face detected as dog Images : {percent} %")

percent = ([dog_detector(x) for x in dog_files_short].count(True))/len(dog_files_short)*100
print(f"Dog detected as dog Face : {percent} %") 
Human face detected as dog Images : 0.0 %
Dog detected as dog Face : 100.0 %

We suggest VGG-16 as a potential network to detect dog images in your algorithm, but you are free to explore other pre-trained networks (such as Inception-v3, ResNet-50, etc). Please use the code cell below to test other pre-trained PyTorch models. If you decide to pursue this optional task, report performance on human_files_short and dog_files_short.

In [13]:
### (Optional) 
### TODO: Report the performance of another pre-trained network.
### Feel free to use as many code cells as needed.

Step 3: Create a CNN to Classify Dog Breeds (from Scratch)

Now that we have functions for detecting humans and dogs in images, we need a way to predict breed from images. In this step, you will create a CNN that classifies dog breeds. You must create your CNN from scratch (so, you can't use transfer learning yet!), and you must attain a test accuracy of at least 10%. In Step 4 of this notebook, you will have the opportunity to use transfer learning to create a CNN that attains greatly improved accuracy.

We mention that the task of assigning breed to dogs from images is considered exceptionally challenging. To see why, consider that even a human would have trouble distinguishing between a Brittany and a Welsh Springer Spaniel.

Brittany Welsh Springer Spaniel

It is not difficult to find other dog breed pairs with minimal inter-class variation (for instance, Curly-Coated Retrievers and American Water Spaniels).

Curly-Coated Retriever American Water Spaniel

Likewise, recall that labradors come in yellow, chocolate, and black. Your vision-based algorithm will have to conquer this high intra-class variation to determine how to classify all of these different shades as the same breed.

Yellow Labrador Chocolate Labrador Black Labrador

We also mention that random chance presents an exceptionally low bar: setting aside the fact that the classes are slightly imabalanced, a random guess will provide a correct answer roughly 1 in 133 times, which corresponds to an accuracy of less than 1%.

Remember that the practice is far ahead of the theory in deep learning. Experiment with many different architectures, and trust your intuition. And, of course, have fun!

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dog_images/train, dog_images/valid, and dog_images/test, respectively). You may find this documentation on custom datasets to be a useful resource. If you are interested in augmenting your training and/or validation data, check out the wide variety of transforms!

In [7]:
import os
from torchvision import datasets, models, transforms
import torch

from PIL import ImageFile
ImageFile.LOAD_TRUNCATED_IMAGES = True


### TODO: Write data loaders for training, validation, and test sets
## Specify appropriate transforms, and batch_sizes
normalize = transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                 std = [0.229, 0.224, 0.225])
data_transforms = {'train': transforms.Compose([transforms.RandomResizedCrop(224),
                                               transforms.RandomHorizontalFlip(), # randomly flip and rotate
                                               transforms.RandomRotation(10),
                                               transforms.ToTensor(),
                                               normalize
                                               ]),
                   'val': transforms.Compose([transforms.Resize((224,224)),
                                               transforms.ToTensor(),
                                               normalize]),
                   'test': transforms.Compose([transforms.Resize((224,224)),
                                               transforms.ToTensor(),
                                               normalize])
                  }


train_data = datasets.ImageFolder('/data/dog_images/train',data_transforms['train'])
valid_data = datasets.ImageFolder('/data/dog_images/valid',data_transforms['val'])
test_data = datasets.ImageFolder('/data/dog_images/test',data_transforms['test'])


print("there total %d number of train images"%len(train_data))
print("there total %d number of train images"%len(valid_data))
print("there total %d number of train images"%len(test_data))

num_workers = 0
batch_size = 20

loaders = {'train':torch.utils.data.DataLoader(train_data,batch_size=batch_size,num_workers=num_workers,shuffle=True),
           'valid': torch.utils.data.DataLoader(valid_data,batch_size=batch_size,num_workers=num_workers,shuffle=True),
           'test':torch.utils.data.DataLoader(test_data,batch_size=batch_size,num_workers=num_workers,shuffle=True)
          }
class_names = train_data.classes
n_classes = len(class_names)
print("the total no of classes:%d"%n_classes)    
there total 6680 number of train images
there total 835 number of train images
there total 836 number of train images
the total no of classes:133
In [3]:
import numpy as np
from matplotlib import colors, cm, pyplot as plt

%matplotlib inline

def imshow(img):
    img = img.transpose((1, 2, 0))
    mean = np.array([0.485, 0.456, 0.406])
    std = np.array([0.229, 0.224, 0.225])
    img = std * img + mean
    img = np.clip(img, 0, 1)    
    plt.imshow(img)

# obtain one batch of training images
dataiter = iter(loaders['train'])
images, labels = dataiter.next()
images = images.numpy() # convert images to numpy for display

# plot the images in the batch, along with the corresponding labels
fig = plt.figure(figsize=(25, 4))
for idx in np.arange(20):
    ax = fig.add_subplot(2, 20/2, idx+1, xticks=[], yticks=[])
    imshow(images[idx])
    ax.set_title(class_names[labels[idx]].split(".")[1])

Question 3: Describe your chosen procedure for preprocessing the data.

  • How does your code resize the images (by cropping, stretching, etc)? What size did you pick for the input tensor, and why?
  • Did you decide to augment the dataset? If so, how (through translations, flips, rotations, etc)? If not, why not?

Answer: 1) Random cropping can also act as a regularizer and base on the classification on the presence of parts of the object instead of focusing everything on a very distinct feature that may not always be present.The original size could be very large to be fed in network hence we crop to a standard 224224.I chose 224 because all the pre-trained models also use this as the input size (3H*W) where , H and W have expected size of 224.

2)Yes I decided to augument the data since it prevents the model to overfit the data by using : a)Random Crop: It is the most common augumentation used since it randomly crops the image and resize it to the given size.I have cropped it 224 size. b)Random Horizontal Flip: It randomly flip some images horizontally hence giving our model to learn the image from differnt propective c)Random Rotation: It randomly rotates the images to a degree mentioned.

(IMPLEMENTATION) Model Architecture

Create a CNN to classify dog breed. Use the template in the code cell below.

In [4]:
import torch.nn as nn
import torch.nn.functional as F

# define the CNN architecture
class Net(nn.Module):
    ### TODO: choose an architecture, and complete the class
    def __init__(self):
        super(Net, self).__init__()
        ## Define layers of a CNN
        self.conv1 = nn.Conv2d( 3, 16, 3,padding=1)
        self.conv2 = nn.Conv2d(16, 32, 3,padding=1)
        self.conv3 = nn.Conv2d(32, 64, 3,padding=1)
        self.conv4 = nn.Conv2d(64, 128, 3,padding=1)
        self.conv5 = nn.Conv2d(128, 256, 3,padding=1)
        # max pooling layer
        self.pool = nn.MaxPool2d(2, 2)
        # linear layer (256 * 5* 5 -> 500)
        self.fc1 = nn.Linear( 256 * 7 * 7, 500)
        self.fc2 = nn.Linear(500, 256)
        self.dropout = nn.Dropout(0.3)

    
    def forward(self, x):
        ## Define forward behavior
        x = self.pool(F.relu(self.conv1(x)))
        x = self.pool(F.relu(self.conv2(x)))
        x = self.pool(F.relu(self.conv3(x)))
        x = self.pool(F.relu(self.conv4(x)))
        x = self.pool(F.relu(self.conv5(x)))
        # flatten image input
        #print(x.shape)
        x = x.view(-1, 7 * 7 * 256)
        # add 1st hidden layer, with relu activation function
        x = self.dropout(x)
        x = F.relu(self.fc1(x))
        # add 2nd hidden layer, with relu activation function
        x = self.dropout(x)
        x = self.fc2(x)
        #x = self.fc3(x)
        return x
#-#-# You so NOT have to modify the code below this line. #-#-#

# instantiate the CNN
model_scratch = Net()

# check if CUDA is available
use_cuda = torch.cuda.is_available()

# move tensors to GPU if CUDA is available
if use_cuda:
    model_scratch.cuda()

Question 4: Outline the steps you took to get to your final CNN architecture and your reasoning at each step.

Answer: I have used five convolution layers as they work as feature extractor I wanted the most features extracted to match a dog so that futher it could map a human too. hence these are my layers :

conv1 : Conv2d(i/p:3 ,o/p:16 ,kernel:3 ,stride:1, padding:1)

pool : Maxpool(size:2 ,stride:2)

conv2 : Conv2d(i/p:16 ,o/p:32 ,kernel:3 ,stride:1, padding:1)

pool : Maxpool(size:2 ,stride:2)

conv3 : Conv2d(i/p:32 ,o/p:64 ,kernel:3 ,stride:1, padding:1)

pool : Maxpool(size:2 ,stride:2)

conv4 : Conv2d(i/p:64 ,o/p:128 ,kernel:3 ,stride:1, padding:1)

pool : Maxpool(size:2 ,stride:2)

conv5 : Conv2d(i/p:128 ,o/p:256 ,kernel:3 ,stride:1, padding:1)

pool : Maxpool(size:2 ,stride:2)

explaination: Due to convolution layer size doesn't change but depth increases and the size is downsized by maxpooling layer with kernel size 2 and stride of 2, which will atlast bring the image to size of 7(*)7 and depth 256 then a dropout of 0.3 will help with overfitting.And lastly it is passed through two fully connected layer (i/p-->hidden) and (hidden-->output) to give out the class results.

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_scratch, and the optimizer as optimizer_scratch below.

In [5]:
import torch.optim as optim

### TODO: select loss function
criterion_scratch = nn.CrossEntropyLoss()

### TODO: select optimizer
optimizer_scratch = optim.Adam(model_scratch.parameters())

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_scratch.pt'.

In [6]:
def train(n_epochs, loaders, model, optimizer, criterion, use_cuda, save_path):
    """returns trained model"""
    # initialize tracker for minimum validation loss
    valid_loss_min = np.Inf 
    
    for epoch in range(1, n_epochs+1):
        # initialize variables to monitor training and validation loss
        train_loss = 0.0
        valid_loss = 0.0
        
        ###################
        # train the model #
        ###################
        model.train()
        for batch_idx, (data, target) in enumerate(loaders['train']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## find the loss and update the model parameters accordingly
            ## record the average training loss, using something like
            ## train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            # clear the gradients of all optimized variables
            optimizer.zero_grad()
            # forward pass: compute predicted outputs by passing inputs to the model
            output = model(data)
            # calculate the batch loss
            loss = criterion(output, target)
            # backward pass: compute gradient of the loss with respect to model parameters
            loss.backward()
            # perform a single optimization step (parameter update)
            optimizer.step()
            # update training loss
            train_loss = train_loss + ((1 / (batch_idx + 1)) * (loss.data - train_loss))
            
        ######################    
        # validate the model #
        ######################
        model.eval()
        for batch_idx, (data, target) in enumerate(loaders['valid']):
            # move to GPU
            if use_cuda:
                data, target = data.cuda(), target.cuda()
            ## update the average validation loss
            output = model(data)
            # calculate the batch loss
            loss = criterion(output, target)
            # update average validation loss 
            valid_loss = valid_loss + ((1 / (batch_idx + 1)) * (loss.data - valid_loss))
            
        train_loss = train_loss/len(loaders['train'])
        valid_loss = valid_loss/len(loaders['valid'])
            
        # print training/validation statistics 
        print('Epoch: {} \tTraining Loss: {:.6f} \tValidation Loss: {:.6f}'.format(
            epoch, 
            train_loss,
            valid_loss
            ))
        
        ## TODO: save the model if validation loss has decreased
        if(valid_loss<=valid_loss_min):
            print('Validation loss decreased ({:.6f} --> {:.6f}).  Saving model ...'.format(valid_loss_min,valid_loss))
            torch.save(model.state_dict(), save_path)
            valid_loss_min = valid_loss
            
    # return trained model
    return model
In [7]:
# train the model
n_epochs = 20
loaders_scratch=loaders
model_scratch = train(n_epochs, loaders_scratch, model_scratch, optimizer_scratch, 
                      criterion_scratch, use_cuda, 'model_scratch.pt')
Epoch: 1 	Training Loss: 0.014883 	Validation Loss: 0.115220
Validation loss decreased (inf --> 0.115220).  Saving model ...
Epoch: 2 	Training Loss: 0.014202 	Validation Loss: 0.109637
Validation loss decreased (0.115220 --> 0.109637).  Saving model ...
Epoch: 3 	Training Loss: 0.013750 	Validation Loss: 0.106223
Validation loss decreased (0.109637 --> 0.106223).  Saving model ...
Epoch: 4 	Training Loss: 0.013519 	Validation Loss: 0.104646
Validation loss decreased (0.106223 --> 0.104646).  Saving model ...
Epoch: 5 	Training Loss: 0.013293 	Validation Loss: 0.102069
Validation loss decreased (0.104646 --> 0.102069).  Saving model ...
Epoch: 6 	Training Loss: 0.013113 	Validation Loss: 0.100991
Validation loss decreased (0.102069 --> 0.100991).  Saving model ...
Epoch: 7 	Training Loss: 0.012839 	Validation Loss: 0.099589
Validation loss decreased (0.100991 --> 0.099589).  Saving model ...
Epoch: 8 	Training Loss: 0.012679 	Validation Loss: 0.097642
Validation loss decreased (0.099589 --> 0.097642).  Saving model ...
Epoch: 9 	Training Loss: 0.012459 	Validation Loss: 0.095293
Validation loss decreased (0.097642 --> 0.095293).  Saving model ...
Epoch: 10 	Training Loss: 0.012331 	Validation Loss: 0.095765
Epoch: 11 	Training Loss: 0.012155 	Validation Loss: 0.095708
Epoch: 12 	Training Loss: 0.012026 	Validation Loss: 0.092686
Validation loss decreased (0.095293 --> 0.092686).  Saving model ...
Epoch: 13 	Training Loss: 0.011851 	Validation Loss: 0.092144
Validation loss decreased (0.092686 --> 0.092144).  Saving model ...
Epoch: 14 	Training Loss: 0.011786 	Validation Loss: 0.090362
Validation loss decreased (0.092144 --> 0.090362).  Saving model ...
Epoch: 15 	Training Loss: 0.011650 	Validation Loss: 0.091440
Epoch: 16 	Training Loss: 0.011597 	Validation Loss: 0.090527
Epoch: 17 	Training Loss: 0.011434 	Validation Loss: 0.089603
Validation loss decreased (0.090362 --> 0.089603).  Saving model ...
Epoch: 18 	Training Loss: 0.011345 	Validation Loss: 0.089088
Validation loss decreased (0.089603 --> 0.089088).  Saving model ...
Epoch: 19 	Training Loss: 0.011255 	Validation Loss: 0.087297
Validation loss decreased (0.089088 --> 0.087297).  Saving model ...
Epoch: 20 	Training Loss: 0.011134 	Validation Loss: 0.085910
Validation loss decreased (0.087297 --> 0.085910).  Saving model ...
In [8]:
# load the model that got the best validation accuracy
model_scratch.load_state_dict(torch.load('model_scratch.pt'))

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 10%.

In [11]:
def test(loaders, model, criterion, use_cuda):

    # monitor test loss and accuracy
    test_loss = 0.
    correct = 0.
    total = 0.

    model.eval()
    for batch_idx, (data, target) in enumerate(loaders['test']):
        # move to GPU
        if use_cuda:
            data, target = data.cuda(), target.cuda()
        # forward pass: compute predicted outputs by passing inputs to the model
        output = model(data)
        # calculate the loss
        loss = criterion(output, target)
        # update average test loss 
        test_loss = test_loss + ((1 / (batch_idx + 1)) * (loss.data - test_loss))
        # convert output probabilities to predicted class
        pred = output.data.max(1, keepdim=True)[1]
        # compare predictions to true label
        correct += np.sum(np.squeeze(pred.eq(target.data.view_as(pred))).cpu().numpy())
        total += data.size(0)
            
    print('Test Loss: {:.6f}\n'.format(test_loss))

    print('\nTest Accuracy: %2d%% (%2d/%2d)' % (
        100. * correct / total, correct, total))
In [10]:
# call test function    
test(loaders_scratch, model_scratch, criterion_scratch, use_cuda)
Test Loss: 3.603817


Test Accuracy: 15% (128/836)

Step 4: Create a CNN to Classify Dog Breeds (using Transfer Learning)

You will now use transfer learning to create a CNN that can identify dog breed from images. Your CNN must attain at least 60% accuracy on the test set.

(IMPLEMENTATION) Specify Data Loaders for the Dog Dataset

Use the code cell below to write three separate data loaders for the training, validation, and test datasets of dog images (located at dogImages/train, dogImages/valid, and dogImages/test, respectively).

If you like, you are welcome to use the same data loaders from the previous step, when you created a CNN from scratch.

In [ ]:
## TODO: Specify data loaders

(IMPLEMENTATION) Model Architecture

Use transfer learning to create a CNN to classify dog breed. Use the code cell below, and save your initialized model as the variable model_transfer.

In [8]:
import torchvision.models as models
import torch.nn as nn

n_classes = 133

## TODO: Specify model architecture 
model_transfer = models.vgg16(pretrained=True)

for params in model_transfer.features.parameters(): # freeze training of all "features" layers 
    params.requires_grad=False
    
print(model_transfer.classifier) 

model_transfer.classifier[6] = nn.Linear(model_transfer.classifier[6].in_features,n_classes)

print(model_transfer.classifier)

# check if CUDA is available
use_cuda = torch.cuda.is_available()

if use_cuda:
    model_transfer = model_transfer.cuda()
Sequential(
  (0): Linear(in_features=25088, out_features=4096, bias=True)
  (1): ReLU(inplace)
  (2): Dropout(p=0.5)
  (3): Linear(in_features=4096, out_features=4096, bias=True)
  (4): ReLU(inplace)
  (5): Dropout(p=0.5)
  (6): Linear(in_features=4096, out_features=1000, bias=True)
)
Sequential(
  (0): Linear(in_features=25088, out_features=4096, bias=True)
  (1): ReLU(inplace)
  (2): Dropout(p=0.5)
  (3): Linear(in_features=4096, out_features=4096, bias=True)
  (4): ReLU(inplace)
  (5): Dropout(p=0.5)
  (6): Linear(in_features=4096, out_features=133, bias=True)
)

Question 5: Outline the steps you took to get to your final CNN architecture and your reasoning at each step. Describe why you think the architecture is suitable for the current problem.

Answer:

I used the VGG16 architecture, pre-trained on the ImageNet dataset. Because the ImageNet dataset contains many "dog" classes among its total of 1000 classes, this model will already have learned features that are relevant to our classification problem.

I then changed the last layer output to fit according to my classes i.e 133

(IMPLEMENTATION) Specify Loss Function and Optimizer

Use the next code cell to specify a loss function and optimizer. Save the chosen loss function as criterion_transfer, and the optimizer as optimizer_transfer below.

In [15]:
from torch import optim
criterion_transfer = nn.CrossEntropyLoss()
optimizer_transfer = optim.SGD(model_transfer.classifier.parameters(),lr=0.001,momentum=0.9)

(IMPLEMENTATION) Train and Validate the Model

Train and validate your model in the code cell below. Save the final model parameters at filepath 'model_transfer.pt'.

In [13]:
# train the model
n_epochs = 25
loaders_transfer = loaders 
model_transfer = train(n_epochs, loaders_transfer, model_transfer, optimizer_transfer, criterion_transfer, use_cuda, 'model_transfer.pt')
Epoch: 1 	Training Loss: 0.006929 	Validation Loss: 0.017568
Validation loss decreased (inf --> 0.017568).  Saving model ...
Epoch: 2 	Training Loss: 0.004136 	Validation Loss: 0.014094
Validation loss decreased (0.017568 --> 0.014094).  Saving model ...
Epoch: 3 	Training Loss: 0.003582 	Validation Loss: 0.012500
Validation loss decreased (0.014094 --> 0.012500).  Saving model ...
Epoch: 4 	Training Loss: 0.003392 	Validation Loss: 0.013003
Epoch: 5 	Training Loss: 0.003338 	Validation Loss: 0.012515
Epoch: 6 	Training Loss: 0.003134 	Validation Loss: 0.012339
Validation loss decreased (0.012500 --> 0.012339).  Saving model ...
Epoch: 7 	Training Loss: 0.003077 	Validation Loss: 0.011802
Validation loss decreased (0.012339 --> 0.011802).  Saving model ...
Epoch: 8 	Training Loss: 0.002891 	Validation Loss: 0.012012
Epoch: 9 	Training Loss: 0.002896 	Validation Loss: 0.011724
Validation loss decreased (0.011802 --> 0.011724).  Saving model ...
Epoch: 10 	Training Loss: 0.002842 	Validation Loss: 0.011639
Validation loss decreased (0.011724 --> 0.011639).  Saving model ...
Epoch: 11 	Training Loss: 0.002582 	Validation Loss: 0.011587
Validation loss decreased (0.011639 --> 0.011587).  Saving model ...
Epoch: 12 	Training Loss: 0.002653 	Validation Loss: 0.010728
Validation loss decreased (0.011587 --> 0.010728).  Saving model ...
Epoch: 13 	Training Loss: 0.002648 	Validation Loss: 0.011522
Epoch: 14 	Training Loss: 0.002589 	Validation Loss: 0.011844
Epoch: 15 	Training Loss: 0.002441 	Validation Loss: 0.011751
Epoch: 16 	Training Loss: 0.002531 	Validation Loss: 0.011071
Epoch: 17 	Training Loss: 0.002494 	Validation Loss: 0.012214
Epoch: 18 	Training Loss: 0.002529 	Validation Loss: 0.011428
Epoch: 19 	Training Loss: 0.002560 	Validation Loss: 0.010951
Epoch: 20 	Training Loss: 0.002454 	Validation Loss: 0.010957
Epoch: 21 	Training Loss: 0.002428 	Validation Loss: 0.010585
Validation loss decreased (0.010728 --> 0.010585).  Saving model ...
Epoch: 22 	Training Loss: 0.002380 	Validation Loss: 0.010960
Epoch: 23 	Training Loss: 0.002239 	Validation Loss: 0.010502
Validation loss decreased (0.010585 --> 0.010502).  Saving model ...
Epoch: 24 	Training Loss: 0.002314 	Validation Loss: 0.011376
Epoch: 25 	Training Loss: 0.002303 	Validation Loss: 0.011507
In [9]:
# load the model that got the best validation accuracy (uncomment the line below)
model_transfer.load_state_dict(torch.load('model_transfer.pt'))

(IMPLEMENTATION) Test the Model

Try out your model on the test dataset of dog images. Use the code cell below to calculate and print the test loss and accuracy. Ensure that your test accuracy is greater than 60%.

In [16]:
loaders_transfer = loaders
test(loaders_transfer, model_transfer, criterion_transfer, use_cuda)
Test Loss: 0.561400


Test Accuracy: 83% (694/836)

(IMPLEMENTATION) Predict Dog Breed with the Model

Write a function that takes an image path as input and returns the dog breed (Affenpinscher, Afghan hound, etc) that is predicted by your model.

In [17]:
### TODO: Write a function that takes a path to an image as input
### and returns the dog breed that is predicted by the model.

# list of class names by index, i.e. a name can be accessed like class_names[0]
class_names = [item[4:].replace("_", " ") for item in loaders['train'].dataset.classes]

def predict_breed_transfer(img_path):
    # load the image and return the predicted breed
    data_transform = transforms.Compose([transforms.Resize(size=(244, 244)),
                                         transforms.ToTensor(),
                                         transforms.Normalize(mean=[0.485, 0.456, 0.406],
                                                             std=[0.229, 0.224, 0.225])])
    
    img_pil = Image.open(img_path).convert('RGB')
    img = data_transform(img_pil).float()
    #img = Variable(img)
    img = img.unsqueeze(0)
    model_transfer.eval()
    if use_cuda:
        img = img.cuda()
    ret = model_transfer(img)
    number = torch.max(ret,1)[1].item()
    
    return class_names[number]
In [18]:
def display_img(img_path):
    img = cv2.imread(img_path)
    cv_rgb = cv2.cvtColor(img, cv2.COLOR_BGR2RGB)
    imgplot = plt.imshow(cv_rgb)
    return imgplot
In [19]:
for img_file in os.listdir('./images'):
    img_path = os.path.join('./images', img_file)
    pic = display_img(img_path)
    plt.show(pic)
    answer = predict_breed_transfer(img_path)
    print("the image is {}\n the dog is {}\n\n".format(img_file,answer))
the image is Curly-coated_retriever_03896.jpg
 the dog is Curly-coated retriever


the image is Labrador_retriever_06457.jpg
 the dog is Labrador retriever


the image is sample_dog_output.png
 the dog is Greyhound


the image is American_water_spaniel_00648.jpg
 the dog is Curly-coated retriever


the image is Labrador_retriever_06455.jpg
 the dog is Labrador retriever


the image is Welsh_springer_spaniel_08203.jpg
 the dog is Welsh springer spaniel


the image is sample_cnn.png
 the dog is Nova scotia duck tolling retriever


the image is Labrador_retriever_06449.jpg
 the dog is Labrador retriever


the image is Brittany_02625.jpg
 the dog is Brittany


the image is sample_human_output.png
 the dog is Lowchen



Step 5: Write your Algorithm

Write an algorithm that accepts a file path to an image and first determines whether the image contains a human, dog, or neither. Then,

  • if a dog is detected in the image, return the predicted breed.
  • if a human is detected in the image, return the resembling dog breed.
  • if neither is detected in the image, provide output that indicates an error.

You are welcome to write your own functions for detecting humans and dogs in images, but feel free to use the face_detector and human_detector functions developed above. You are required to use your CNN from Step 4 to predict dog breed.

Some sample output for our algorithm is provided below, but feel free to design your own user experience!

Sample Human Output

(IMPLEMENTATION) Write your Algorithm

In [22]:
### TODO: Write your algorithm.
### Feel free to use as many code cells as needed.

def run_app(img_path):
    pic = display_img(img_path)
    plt.show(pic)
    ## handle cases for a human face, dog, and neither
    if face_detector(img_path):
        prediction = predict_breed_transfer(img_path)
        print("hey this looks like a human!!\n but to prdict a dog type would be{}".format(prediction))
    elif dog_detector(img_path):
        prediction = predict_breed_transfer(img_path)
        print("hey this looks like a dog and its type would be {}".format(prediction))
    else:
        print("this is neither a dog or a human sorry !! no prediction ")
        
In [18]:
for img_file in os.listdir('./images'):
    img_path = os.path.join('./images', img_file)
    run_app(img_path)
hey this looks like a dog and its type would be Chow chow
hey this looks like a dog and its type would be Bernese mountain dog
hey this looks like a dog and its type would be Pharaoh hound
this is neither a dog or a human sorry !! no prediction 
hey this looks like a dog and its type would be Tibetan mastiff
hey this looks like a dog and its type would be Dandie dinmont terrier
hey this looks like a dog and its type would be Bichon frise
hey this looks like a human!!
 but to prdict a dog type would bePointer
hey this looks like a dog and its type would be Belgian malinois
hey this looks like a dog and its type would be English setter

Step 6: Test Your Algorithm

In this section, you will take your new algorithm for a spin! What kind of dog does the algorithm think that you look like? If you have a dog, does it predict your dog's breed accurately? If you have a cat, does it mistakenly think that your cat is a dog?

(IMPLEMENTATION) Test Your Algorithm on Sample Images!

Test your algorithm at least six images on your computer. Feel free to use any images you like. Use at least two human and two dog images.

Question 6: Is the output better than you expected :) ? Or worse :( ? Provide at least three possible points of improvement for your algorithm.

Answer: (Three possible points for improvement)

The output is better that I expected but when classifiying people as dog type for same person in two different images it is giving me different dog types. This could be either be assumed as noise or we need to train better on feature extraction.

1)more Hyper-parameter tunings could be explored, differnt agumentation has shown good results

2)The dataset being large should be able to give more accurate results (more breed types )

3)other model architectures can be explored to reduce raining time without comprimising on accuracy

In [20]:
human_files = ['./my_images/human_1.jpg', './my_images/human_2.jpg' ]
dog_files = ['./my_images/pug.jpg', './my_images/husky.jpg', './my_images/lab.jpg']
In [23]:
## TODO: Execute your algorithm from Step 6 on
## at least 6 images on your computer.
## Feel free to use as many code cells as needed.

## suggested code, below
for file in np.hstack((human_files, dog_files)):
    run_app(file)
hey this looks like a human!!
 but to prdict a dog type would beMastiff
hey this looks like a human!!
 but to prdict a dog type would beEnglish springer spaniel
hey this looks like a dog and its type would be French bulldog
hey this looks like a dog and its type would be Icelandic sheepdog
hey this looks like a dog and its type would be Labrador retriever
In [ ]: